
# Librerías importantes

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from keras.preprocessing.image import ImageDataGenerator
from keras.models import Sequential
from keras.layers import Dropout, Flatten, Dense
from keras import applications
from sklearn.metrics import pairwise_distances
import requests
from PIL import Image
import pickle
from datetime import datetime
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
import plotly.figure_factory as ff
import plotly.graph_objects as go
import plotly.express as px
import streamlit as st
#use the below library while displaying the images in jupyter notebook
from IPython.display import display, Image
import streamlit as st

houses_df = pd.read_excel("C:\\Users\\alexa\\OneDrive\\CONCENTRACION_LIT\\RETOFASE2\\houses.xlsx") # Set de datos con los id y categorias de los objetos
houses_df


houses_df["Style"].value_counts()
print("Numero total de casas: ", houses_df.shape[0])
print("Numero de estilos de casas: ", houses_df["Style"].nunique())

Numero total de casas:  91
Numero de estilos de casas:  4


# Se tienen 4 estilos
houses_contemporary = houses_df[houses_df["Style"]=="Contemporary/Modern"]
houses_ranch = houses_df[houses_df["Style"]=="Ranch"]
houses_traditional = houses_df[houses_df["Style"]=="Traditional"]
houses_mediterranean = houses_df[houses_df["Style"]=="Mediterranean"]


# Directorios (Cambiar por el propio cuando se descarguen las bases)
path = "C:\\Users\\alexa\\OneDrive\\CONCENTRACION_LIT\\RETOFASE2" # Directorio general donde se depositaran las caracteristicas e indices de las imagenes
try:
   os.chdir(path)
   print("Directorio actual: {0}".format(os.getcwd()))
except FileNotFoundError:
   print("Directorio: {0} no existe".format(path))
except NotADirectoryError:
   print("{0} no es un directorio".format(path))
except PermissionError:
   print("Sin permiso de modificar {0}".format(path))

Directorio actual: C:\Users\alexa\OneDrive\CONCENTRACION_LIT\RETOFASE2


img_width, img_height = 224, 224 # Pixeles de imagenes
train_data_dir = "C:\\Users\\alexa\\OneDrive\\CONCENTRACION_LIT\\RETOFASE2\\HOUSES_IMAGES\\CONT" # Directorio de imagenes de contemporary houses


train_data_dir

'C:\\Users\\alexa\\OneDrive\\CONCENTRACION_LIT\\RETOFASE2\\HOUSES_IMAGES\\CONT'


nb_train_samples = 20  # Muestras
epochs = 100  # Epocas (Episodios)
batch_size = 1  # Tamaño del lote

# Nota: Variar los parámetros de la parte superior (muestras, episodios y tamaño del lote) para modificar el rendimiento de la red

def extract_features():
    Itemcodes = []
    datagen = ImageDataGenerator(rescale=1. / 255)  # Escalamiento de las imagenes
    model = applications.ResNet50(include_top=False, weights='imagenet')  # Uso de ResNet50
    generator = datagen.flow_from_directory(
        train_data_dir,
        target_size=(img_width, img_height),
        batch_size=batch_size,
        class_mode=None,
        shuffle=False
    )
    for i in generator.filenames:
        Itemcodes.append(i[(i.find("/") + 1):i.find(".")])
    extracted_features = model.predict(generator, nb_train_samples // batch_size)  # Extracción de características

    num_samples = extracted_features.shape[0]  # Get the actual number of samples
    num_features = extracted_features.size // num_samples  # Calculate the number of features per sample
    extracted_features = extracted_features.reshape((num_samples, -1))  # Reshape into (num_samples, num_features)

    np.save(open('./CONTEMPORARY_ResNet_features.npy', 'wb'), extracted_features)

a = datetime.now()
extract_features()
print("Tiempo de extraccion de caracteristicas", datetime.now() - a)

Found 20 images belonging to 1 classes.
20/20 [==============================] - 3s 96ms/step
Tiempo de extraccion de caracteristicas 0:00:05.718148


extracted_features = np.load('CONTEMPORARY_ResNet_features.npy')
contemporary =houses_contemporary.copy()
#contemporary = pd.read_csv('./houses_contemporary.csv')
df_Productids = list(contemporary['HouseId']) # Id de los productos


def get_similar_products_cnn(product_id, num_results):
    doc_id = df_Productids.index(product_id)
    pairwise_dist = pairwise_distances(extracted_features, extracted_features[doc_id].reshape(1,-1))
    indices = np.argsort(pairwise_dist.flatten())[0:num_results]
    pdists  = np.sort(pairwise_dist.flatten())[0:num_results]
    print("="*20, "House Image", "="*20)
    ip_row = contemporary[['ImageURL','HouseTitle']].loc[contemporary['HouseId']==int(df_Productids[indices[0]])]
    #print(ip_row.head())
    for indx, row in ip_row.iterrows():
        display(Image(url=row['ImageURL'], width = 224, height = 224,embed=True))
        print('House Title: ', row['HouseTitle'])
    print("\n","="*20, "Recommended Houses", "="*20)
    for i in range(1,len(indices)):
        rows = contemporary[['ImageURL','HouseTitle']].loc[contemporary['HouseId']==int(df_Productids[indices[i]])]
        for indx, row in rows.iterrows():
            display(Image(url=row['ImageURL'], width = 224, height = 224,embed=True))
            print('House Title: ', row['HouseTitle'])
            print('Distancia euclidiana desde la imagen muestra:', pdists[i])


En este caso como ejemplo tomamos la casa 50633300 del tipo contemporaneo y pedimos que muestre las 5 casas mas parecidas a 
la misma.


get_similar_products_cnn(50633300, 5)

==================== House Image ====================

House Title:  5 Crinkleroot Ct

 ==================== Recommended Houses ====================

House Title:  139 Vue Point Pl
Distancia euclidiana desde la imagen muestra: 35.923786

House Title:  39 Thornblade Cir
Distancia euclidiana desde la imagen muestra: 35.985115

House Title:   19 E Sunny Slope Cir
Distancia euclidiana desde la imagen muestra: 38.81897

House Title:   79 Huntsmans Horn Cir
Distancia euclidiana desde la imagen muestra: 38.902916


img_width, img_height = 224, 224


#top_model_weights_path = 'resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5'
train_data_dir = "C:\\Users\\alexa\\OneDrive\\CONCENTRACION_LIT\\RETOFASE2\\HOUSES_IMAGES\\RAN" # Directorio de imagenes de mujeres

# Nota: Se aplica el mismo procedimiento que el usado en las imagenes de ropa de hombre

nb_train_samples = 25 # Muestras CAMBIAR
epochs = 100  # Epocas (Episodios)
batch_size = 1  # Tamaño del lote

# Nota: Variar los parámetros de la parte superior (muestras, episodios y tamaño del lote) para modificar el rendimiento de la red

def extract_features():
    Itemcodes = []
    datagen = ImageDataGenerator(rescale=1. / 255)  # Escalamiento de las imagenes
    model = applications.ResNet50(include_top=False, weights='imagenet')  # Uso de ResNet50
    generator = datagen.flow_from_directory(
        train_data_dir,
        target_size=(img_width, img_height),
        batch_size=batch_size,
        class_mode=None,
        shuffle=False
    )
    for i in generator.filenames:
        Itemcodes.append(i[(i.find("/") + 1):i.find(".")])
    extracted_features = model.predict(generator, nb_train_samples // batch_size)  # Extracción de características

    num_samples = extracted_features.shape[0]  # Get the actual number of samples
    num_features = extracted_features.size // num_samples  # Calculate the number of features per sample
    extracted_features = extracted_features.reshape((num_samples, -1))  # Reshape into (num_samples, num_features)

    np.save(open('./RANCH_ResNet_features.npy', 'wb'), extracted_features)

a = datetime.now()
extract_features()
print("Tiempo de extraccion de caracteristicas", datetime.now() - a)

Found 25 images belonging to 1 classes.
25/25 [==============================] - 3s 91ms/step
Tiempo de extraccion de caracteristicas 0:00:05.163063


extracted_features = np.load('RANCH_ResNet_features.npy')
ranch =houses_ranch.copy()
#contemporary = pd.read_csv('./houses_contemporary.csv')
df_Productids = list(ranch['HouseId']) # Id de los productos


def get_similar_products_cnn(product_id, num_results):
    doc_id = df_Productids.index(product_id)
    pairwise_dist = pairwise_distances(extracted_features, extracted_features[doc_id].reshape(1,-1))
    indices = np.argsort(pairwise_dist.flatten())[0:num_results]
    pdists  = np.sort(pairwise_dist.flatten())[0:num_results]
    print("="*20, "House Image", "="*20)
    ip_row = ranch[['ImageURL','HouseTitle']].loc[ranch['HouseId']==int(df_Productids[indices[0]])]
    #print(ip_row.head())
    for indx, row in ip_row.iterrows():
        display(Image(url=row['ImageURL'], width = 224, height = 224,embed=True))
        print('House Title: ', row['HouseTitle'])
    print("\n","="*20, "Recommended Houses", "="*20)
    for i in range(1,len(indices)):
        rows = ranch[['ImageURL','HouseTitle']].loc[ranch['HouseId']==int(df_Productids[indices[i]])]
        for indx, row in rows.iterrows():
            display(Image(url=row['ImageURL'], width = 224, height = 224,embed=True))
            print('House Title: ', row['HouseTitle'])
            print('Distancia euclidiana desde la imagen muestra:', pdists[i])


get_similar_products_cnn(10901549, 5)

==================== House Image ====================

House Title:  14 Camber Pine Pl

 ==================== Recommended Houses ====================

House Title:  2 Tangle Brush Dr
Distancia euclidiana desde la imagen muestra: 23.89868

House Title:  37 Lance Leaf Road
Distancia euclidiana desde la imagen muestra: 25.358463

House Title:  21 E Brookberry Ct
Distancia euclidiana desde la imagen muestra: 25.419004

House Title:  5 Terravale Court
Distancia euclidiana desde la imagen muestra: 25.526335


img_width, img_height = 224, 224


#top_model_weights_path = 'resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5'
train_data_dir = "C:\\Users\\alexa\\OneDrive\\CONCENTRACION_LIT\\RETOFASE2\\HOUSES_IMAGES\\TRADI" # Directorio de imagenes de mujeres

# Nota: Se aplica el mismo procedimiento que el usado en las imagenes de ropa de hombre

nb_train_samples = 20 # Muestras CAMBIAR
epochs = 100  # Epocas (Episodios)
batch_size = 1  # Tamaño del lote

# Nota: Variar los parámetros de la parte superior (muestras, episodios y tamaño del lote) para modificar el rendimiento de la red

def extract_features():
    Itemcodes = []
    datagen = ImageDataGenerator(rescale=1. / 255)  # Escalamiento de las imagenes
    model = applications.ResNet50(include_top=False, weights='imagenet')  # Uso de ResNet50
    generator = datagen.flow_from_directory(
        train_data_dir,
        target_size=(img_width, img_height),
        batch_size=batch_size,
        class_mode=None,
        shuffle=False
    )
    for i in generator.filenames:
        Itemcodes.append(i[(i.find("/") + 1):i.find(".")])
    extracted_features = model.predict(generator, nb_train_samples // batch_size)  # Extracción de características

    num_samples = extracted_features.shape[0]  # Get the actual number of samples
    num_features = extracted_features.size // num_samples  # Calculate the number of features per sample
    extracted_features = extracted_features.reshape((num_samples, -1))  # Reshape into (num_samples, num_features)

    np.save(open('./TRADITIONAL_ResNet_features.npy', 'wb'), extracted_features)

a = datetime.now()
extract_features()
print("Tiempo de extraccion de caracteristicas", datetime.now() - a)

Found 20 images belonging to 1 classes.
20/20 [==============================] - 3s 109ms/step
Tiempo de extraccion de caracteristicas 0:00:05.636039


extracted_features = np.load('TRADITIONAL_ResNet_features.npy')
tradi =houses_traditional.copy()
#contemporary = pd.read_csv('./houses_contemporary.csv')
df_Productids = list(tradi['HouseId']) # Id de los productos


def get_similar_products_cnn(product_id, num_results):
    doc_id = df_Productids.index(product_id)
    pairwise_dist = pairwise_distances(extracted_features, extracted_features[doc_id].reshape(1,-1))
    indices = np.argsort(pairwise_dist.flatten())[0:num_results]
    pdists  = np.sort(pairwise_dist.flatten())[0:num_results]
    print("="*20, "House Image", "="*20)
    ip_row = tradi[['ImageURL','HouseTitle']].loc[tradi['HouseId']==int(df_Productids[indices[0]])]
    #print(ip_row.head())
    for indx, row in ip_row.iterrows():
        display(Image(url=row['ImageURL'], width = 224, height = 224,embed=True))
        print('House Title: ', row['HouseTitle'])
    print("\n","="*20, "Recommended Houses", "="*20)
    for i in range(1,len(indices)):
        rows = tradi[['ImageURL','HouseTitle']].loc[tradi['HouseId']==int(df_Productids[indices[i]])]
        for indx, row in rows.iterrows():
            display(Image(url=row['ImageURL'], width = 224, height = 224,embed=True))
            print('House Title: ', row['HouseTitle'])
            print('Distancia euclidiana desde la imagen muestra:', pdists[i])


get_similar_products_cnn(9149926, 5)

==================== House Image ====================

House Title:  91 E Bracebridge Cir

 ==================== Recommended Houses ====================

House Title:  11 Tiburon Ct
Distancia euclidiana desde la imagen muestra: 46.40416

House Title:  238 Flaxen Chestnut
Distancia euclidiana desde la imagen muestra: 49.336792

House Title:  110 E Grey Wing Court Ct
Distancia euclidiana desde la imagen muestra: 49.496166

House Title:  1 Southgate Dr
Distancia euclidiana desde la imagen muestra: 49.735847


img_width, img_height = 224, 224


#top_model_weights_path = 'resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5'
train_data_dir = "C:\\Users\\alexa\\OneDrive\\CONCENTRACION_LIT\\RETOFASE2\\HOUSES_IMAGES\\MEDI" # Directorio de imagenes de mujeres

# Nota: Se aplica el mismo procedimiento que el usado en las imagenes de ropa de hombre

nb_train_samples = 26  # Muestras CAMBIAR
epochs = 100  # Epocas (Episodios)
batch_size = 1  # Tamaño del lote

# Nota: Variar los parámetros de la parte superior (muestras, episodios y tamaño del lote) para modificar el rendimiento de la red

def extract_features():
    Itemcodes = []
    datagen = ImageDataGenerator(rescale=1. / 255)  # Escalamiento de las imagenes
    model = applications.ResNet50(include_top=False, weights='imagenet')  # Uso de ResNet50
    generator = datagen.flow_from_directory(
        train_data_dir,
        target_size=(img_width, img_height),
        batch_size=batch_size,
        class_mode=None,
        shuffle=False
    )
    for i in generator.filenames:
        Itemcodes.append(i[(i.find("/") + 1):i.find(".")])
    extracted_features = model.predict(generator, nb_train_samples // batch_size)  # Extracción de características

    num_samples = extracted_features.shape[0]  # Get the actual number of samples
    num_features = extracted_features.size // num_samples  # Calculate the number of features per sample
    extracted_features = extracted_features.reshape((num_samples, -1))  # Reshape into (num_samples, num_features)

    np.save(open('./MEDITERRANEAN_ResNet_features.npy', 'wb'), extracted_features)

a = datetime.now()
extract_features()
print("Tiempo de extraccion de caracteristicas", datetime.now() - a)

Found 26 images belonging to 1 classes.
26/26 [==============================] - 4s 107ms/step
Tiempo de extraccion de caracteristicas 0:00:06.699287


extracted_features = np.load('MEDITERRANEAN_ResNet_features.npy')
medi =houses_mediterranean.copy()
#contemporary = pd.read_csv('./houses_contemporary.csv')
df_Productids = list(medi['HouseId']) # Id de los productos


def get_similar_products_cnn(product_id, num_results):
    doc_id = df_Productids.index(product_id)
    pairwise_dist = pairwise_distances(extracted_features, extracted_features[doc_id].reshape(1,-1))
    indices = np.argsort(pairwise_dist.flatten())[0:num_results]
    pdists  = np.sort(pairwise_dist.flatten())[0:num_results]
    print("="*20, "House Image", "="*20)
    ip_row = medi[['ImageURL','HouseTitle']].loc[medi['HouseId']==int(df_Productids[indices[0]])]
    #print(ip_row.head())
    for indx, row in ip_row.iterrows():
        display(Image(url=row['ImageURL'], width = 224, height = 224,embed=True))
        print('House Title: ', row['HouseTitle'])
    print("\n","="*20, "Recommended Houses", "="*20)
    for i in range(1,len(indices)):
        rows = medi[['ImageURL','HouseTitle']].loc[medi['HouseId']==int(df_Productids[indices[i]])]
        for indx, row in rows.iterrows():
            display(Image(url=row['ImageURL'], width = 224, height = 224,embed=True))
            print('House Title: ', row['HouseTitle'])
            print('Distancia euclidiana desde la imagen muestra:', pdists[i])


get_similar_products_cnn(3108527, 5)

==================== House Image ====================

House Title:  83 Lakeside Grn

 ==================== Recommended Houses ====================

House Title:  51 Ironton Place
Distancia euclidiana desde la imagen muestra: 27.289204

House Title:  11 Papado Trails Circle
Distancia euclidiana desde la imagen muestra: 28.980757

House Title:  6 Paloma Bend Place
Distancia euclidiana desde la imagen muestra: 29.419434

House Title:  35 Winter Thicket Place
Distancia euclidiana desde la imagen muestra: 29.889215


import numpy as np
import pandas as pd
from PIL import Image
from PIL import ImageFile
import urllib.request
#from IPython.display import display, Image
from sklearn.metrics import pairwise_distances
from datetime import datetime
import streamlit as st

# Directorios (Cambiar por el propio cuando se descarguen las bases)
path = "C:/Users/alexa/OneDrive/CONCENTRACION_LIT/RETOFASE2"

 # Directorio general donde se depositaran las caracteristicas e indices de las imagenes
try:
   os.chdir(path)
   print("Directorio actual: {0}".format(os.getcwd()))
except FileNotFoundError:
   print("Directorio: {0} no existe".format(path))
except NotADirectoryError:
   print("{0} no es un directorio".format(path))
except PermissionError:
   print("Sin permiso de modificar {0}".format(path))

# In[4]:

st.set_option('deprecation.showfileUploaderEncoding', False)

# Uso de las características extraídas por categoría

houses_df = pd.read_excel("C:/Users/alexa/OneDrive/CONCENTRACION_LIT/RETOFASE2/houses.xlsx")
contemporary_features = np.load('C:/Users/alexa/OneDrive/CONCENTRACION_LIT/RETOFASE2/CONTEMPORARY_ResNet_features.npy')
mediterranean_features = np.load('C:/Users/alexa/OneDrive/CONCENTRACION_LIT/RETOFASE2/MEDITERRANEAN_ResNet_features.npy')
traditional_features = np.load('C:/Users/alexa/OneDrive/CONCENTRACION_LIT/RETOFASE2/TRADITIONAL_ResNet_features.npy')
ranch_features = np.load('C:/Users/alexa/OneDrive/CONCENTRACION_LIT/RETOFASE2/RANCH_ResNet_features.npy')
houses_df["HouseId"] = houses_df["HouseId"].astype(str)

# Indices de los objetos por categoría


houses_contemporary = houses_df[houses_df["Style"]=="Contemporary/Modern"]
houses_ranch = houses_df[houses_df["Style"]=="Ranch"]
houses_traditional = houses_df[houses_df["Style"]=="Traditional"]
houses_mediterranean = houses_df[houses_df["Style"]=="Mediterranean"]

def get_similar_products_cnn(product_id, num_results): # Primero definir la categoria
    if(houses_df[houses_df['HouseId']==product_id]['Style'].values[0]=="Contemporary/Modern"):
        extracted_features = contemporary_features
        Productids = list(houses_contemporary['HouseId'])
    elif(houses_df[houses_df['HouseId']==product_id]['Style'].values[0]=="Ranch"):
        extracted_features = ranch_features
        Productids = list(houses_ranch['HouseId'])
    elif(houses_df[houses_df['HouseId']==product_id]['Style'].values[0]=="Mediterranean"):
        extracted_features = mediterranean_features
        Productids = list(houses_mediterranean['HouseId'])
    elif(houses_df[houses_df['HouseId']==product_id]['Style'].values[0]=="Traditional"):
        extracted_features = traditional_features
        Productids = list(houses_traditional['HouseId'])
    Productids = list(Productids)
    doc_id = Productids.index(product_id) # Id de los productos
    pairwise_dist = pairwise_distances(extracted_features, extracted_features[doc_id].reshape(1,-1))
    indices = np.argsort(pairwise_dist.flatten())[0:num_results]
    pdists  = np.sort(pairwise_dist.flatten())[0:num_results]
    st.write("""
         #### Details 
         """)
    ip_row = houses_df[['ImageURL','HouseTitle']].loc[houses_df['HouseId']==Productids[indices[0]]]
    for indx, row in ip_row.iterrows():
        image = Image.open(urllib.request.urlopen(row['ImageURL']))
        image = image.resize((224,224))
        st.image(image)
        st.write(f"Chosen House : {row['HouseTitle']}")
    st.write(f"""
         #### The {num_results} houses recommended for you are: 
         """)
    for i in range(1,len(indices)):
        rows = houses_df[['ImageURL','HouseTitle']].loc[houses_df['HouseId']==Productids[indices[i]]]
        for indx, row in rows.iterrows():
            #image = Image.open(Image(url=row['ImageURL'], width = 224, height = 224,embed=True))
            image = Image.open(urllib.request.urlopen(row['ImageURL']))
            image = image.resize((224,224))
            st.image(image)
            st.write(f"House Title: {row['HouseTitle']}")
            st.write(f"Distance of the image: {pdists[i]}")

st.write("""
         ## House Recommendation
         """
         )


user_input1 = st.text_input("Insert the Id of the house you liked the most")
user_input2 = st.text_input("Insert the number of recommendations you would like to see")


button = st.button('Generate recommendations')
if button:
    get_similar_products_cnn(str(user_input1), int(user_input2))


from IPython.display import Image
Image(filename='C:/Users/alexa/OneDrive/CONCENTRACION_LIT/RETOFASE2/recon.jpeg')

	HouseId	Type	Sub_Type	Style	Price	HouseTitle	Image	ImageURL
0	6740504	Buy	Single-Family	Contemporary/Modern	2500000	114 S Timber Top Dr	6740504.jpg	https://rew-feed-images.global.ssl.fastly.net/...
1	14498187	Buy	Townhouse Condominium	Contemporary/Modern	269000	18 S Villa Oaks Dr	14498187.jpg	https://rew-feed-images.global.ssl.fastly.net/...
2	18045693	Buy	Single-Family	Contemporary/Modern	550000	3 N Timber Top Dr	18045693.jpg	https://rew-feed-images.global.ssl.fastly.net/...
3	22311505	Buy	Single-Family	Contemporary/Modern	2290000	139 Vue Point Pl	22311505.jpg	https://rew-feed-images.global.ssl.fastly.net/...
4	27593002	Rental	Single Family Detached	Contemporary/Modern	2750	2 Eagle Rock Pl	27593002.jpg	https://rew-feed-images.global.ssl.fastly.net/...
...	...	...	...	...	...	...	...	...
86	95801532	Sold	Single-Family	Mediterranean	1275000	51 Ironton Place	95801532.jpg	https://ap.rdcpix.com/ac91e802cffb72613f6fa09c...
87	28451041	Sold	Single-Family	Mediterranean	763229	11 Papado Trails Circle	28451041.jpg	https://ap.rdcpix.com/1949bc7460173a1912098136...
88	57585680	Sold	Single-Family	Mediterranean	598000	29 Cokeberry Street	57585680.jpg	https://ssl.cdn-redfin.com/photo/104/mbpaddedw...
89	37286582	Sold	Single-Family	Mediterranean	967725	17 Honey Daffodil Place	37286582.jpg	https://photos.zillowstatic.com/fp/54d11015cbf...
90	45040581	Sold	Single-Family	Mediterranean	1050000	23 Olmstead Row	45040581.jpg	https://photos.harstatic.com/191763015/hr/img-...

Modelo Datos No Estructurados¶

Reconocimiento de imagenes¶

Observando caracteristicas y renombrando estilos¶

Definiendo el numero de estilos¶

Estilo Modern/Contemporary¶

Categorías restantes¶

RANCH¶

TRADITIONAL¶

MEDITERRANEAN¶

Visual Studio Code¶

Ejecución¶

Visualización¶